IS SYNAESTHESIA ACTUALLY IDEAESTESIA?

AN INQUIRY INTO THE

NATURE OF THE PHENOMENON
Danko Nikoli
Max-Planck Institute for Brain Research, Frankfurt/M., Germany
Frankfurt Institute for Advanced Studies, Frankfurt/M., Germany
danko.nikolic@gmail.com

Abstract: Synaesthesia is traditionally described as a phenomenon of intermixed

senses. This implies that both, the inducer and concurrent operate at the level of their
sensory representations. For example, in the case of grapheme-colour synaesthesia, the
sensory representation of one type, that of a grapheme, would induce a sensory
representation of another type, a colour. This sensory-sensory view of synaesthesia
has a long tradition as it is embedded also into the very name of the phenomenon:
syn+aisthesis (in Greek: unity of senses). However, evidence has accumulated
suggesting that we should break out with this tradition and adopt a different view of the
phenomenon. In this view, only the concurrent operates at the sensory level of
representation. The inducer, in contrast, contributes from the semantic level of
representationsa processing stage at which the meaning of the stimulus is extracted
and represented. Therefore, synaesthesia can be understood as an unusual type of a
semantic association whereby, in addition to wiring up different concepts,
synaesthesia wires concepts to sensory activations. Thus, a more accurate name of the
phenomenon is ideaesthesia, coined from idea+aisthesis (in Greek: sensing
concepts).
Keywords: synaesthesia, semantics, meaning, concepts, inducer, definition, Glagolitic
letters, ideaesthesia,

There is much, objective, third-person evidence that

synaesthetic experiences are vivid and realistic
(NUNN, et al., 2002; HUBBARD, et al. 2005; NIKOLI,
et al., 2007a). Thus, the sensory or perceptual
nature of concurrents is supported well by
experimental investigations. The situation is less
clear with the inducers. Traditionally, it has been
assumed that inducers operate in synaesthesia at the
sensory level too. However, the idea of the sensory
nature of inducer's role in synaesthesia is not
supported by experimental evidence. A number of
experiments suggest that it is the activation of the
meaning of a particular stimulus that induces
synaesthetic experiences.
For example, one study investigated synaesthetic
perceptions when subjects were presented with a
physically identical grapheme but, depending on the
context, were lead to believe that the grapheme
indicated either a digit zero or a letter (e.g. a
grapheme can be made to be about equally similar to

a number 5 or to an S). The results showed that

the synaesthetic colors associated with the grapheme
changed reliably in dependence on the assumed
meaning of the grapheme (DIXON, et al., 2006)1. As
the stimulus remained constant, the only variable that
changed in these studies was the interpretation of the
grapheme, i.e. its understood meaning. Therefore, the
semantic representation was the factor that
determined the associated synaesthetic color. There
are also other studies from which similar conclusions
can be derived. For example, a stimulus indicating
digit 5 can be build from elements that look like
digit 2. Depending on whether the attention is
focused on the elements or on the whole figure,
different concepts are evoked despite the stimulus
remaining physically identical (PALMERI, et al.,
2002). Here, again, a synaesthete subject perceived
changes in the associated color consistent with the
1
In my lab, we were able to obtain similar results by using
a circular grapheme to represent either a zero or letter O
(unpublished observation).

semantic hypothesis of the nature of the induceri.e.

depending on the interpretation of the stimulus. There
were also other studies whose results suggest
semantic nature of inducers (DIXON, et al. 2000;
RICH, MATTINGLEY, 2003; SIMNER, WARD, 2006),
although not always did the authors offer these
interpretations.
We recently tested the semantic hypothesis by
attempting to induce new associations between the
inducers and concurrents. We reasoned that, if the
inducer operated at the level of semantic
representations, a novel synaesthetic association
between a new grapheme could be established
quickly, i.e. within minutes, by simply associating an
old grapheme with a new one. We taught subjects
ancient unfamiliar Glagolitic alphabet. For example,
if a synaesthete had a red color associated with Latin
letter A and learned a Glagolitic grapheme
standing for A, a presentation of the Glagolitic
grapheme alone would be sufficient to induce
synaesthetic colors previously associated to Latin
A. Thus, synaesthetic colors were transferred as
soon as the subjects learned the meaning of the new
graphemes. Importantly, this transfer was very fast,
requiring <10 minutes of a learning exercise
(MROCZKO, et al., 2008). We could also prove the
existence of the novel association by applying a
Stroop task adapted for synaesthesia. This indicated
semantic
nature
of
novel
graphemecolor
associations.
The conclusions about the inducer's semantic nature
in synaesthesia were challenged most strongly by the
experiments based on serial visual search. One result
apparently suggested a different conclusion, i.e. that
synaesthetic inducers operate at the level of
perception (referred to also as the sensory level).
The idea was that, if graphemes begin to induce
colors at this low level of representation, synaesthetes
should be faster in serial search tasks than the control
subjects. This experiment required the target and
distractor items to be made of different graphemes,
each being associated with a different color. A
synaesthete subject was found to be much faster in
serial search task than a group of non-synaesthete
control subjects. This result was taken to support the
existence of early, pre-attentive, and perceptionrelated associations between the form of the
grapheme and the color. The problem was however,
that this study could not be replicated (EDQUIST, et
al., 2006; SAGIV, et al., 2006). In my own laboratory,
we also attempted and filed to replicate these result
(NIKOLI, et al., 2007b). Our synaesthetes were not
faster in visual search that the matched controls.
Thus, after all, the hypothesis of the low-level
association between the inducer and the concurrent
has been falsified. Consequently, also the idea that
synaesthesia occurs through unification of senses can
be considered falsified.
There has been one other relevant study, a derivative
from PALMERI, et al. (2002), which used not one, but

multiple targets in a serial search task, and subjects

were instructed to detect the shape formed by the
targets rather than detecting a single target
(RAMACHANDRAN, HUBBARD, 2001). In this study,
synaesthetes performed better than controls, and
hence, the results were interpreted as supporting the
hypothesis
of
low-level
representation
of
synaesthesia. However, as mentioned earlier, the
perceptual hypothesis has been in the meantime
falsified on the basis of the experiments with a single
target. Thus, synaesthetes' advantage in tasks with
multiple but not in those with single targets can
hardly be used as a support for the perceptual nature
of inducers. Instead, the reconciliation between the
results of RAMACHANDRAN & HUBBARD and those
from the serial search experiments has to come from
an alternative interpretation of the result of
RAMACHANDRAN & HUBBARD. One possibility is
that the more complex task, with multiple targets,
engages also the semantic networks and that this is
the cause of the advantage of synaesthetes. This
possibility is suggested by recent studies showing
that, with multiple targets, in addition to the detection
the targets there is another time-consuming process
for learning the locations of these target elements
(NIKOLI, SINGER, 2007; MAYER, et al., 2007). The
later relies on limited-capacity resources of visual
working memory and attention, and, most
importantly, the effective working memory capacity,
and hence the learning efficacy, depend crucially on
perceptual conditions (pop-out vs. no pop-out)
(NIKOLI, SINGER, 2007). More familiar shapes are
expected to be detected faster. Thus, synaesthetic
colors may enhance the process of learning the
shapes, rather than the detection of the constituent
elements, the initial element-detection being executed
always slowly but the integration into the shape being
facilitated by the synaesthetic colors. In that way,
synaesthesia would assist the formation of the
perceptual memories of the shapes, which would in
turn facilitate later the detection of these shapes in
crowded scenes. This hypothesis needs, however, yet
to be investigated.
Nevertheless, when put on scales, the evidence for
perceptual and semantic hypothesis overweighs
convincingly towards the conclusion that the inducers
in synaesthesia evoke synaesthetic associations from
the higher-semantic levels of representation. There is
lack of evidence that inducers operate at the low,
perceptual levels of representation. Hence, the events
underlying synaesthesia begin unfolding most likely
with recognition and classification of each stimulus
according to its meaning, the concurrent experiences
being induced only in a subsequent stage. In other
words, it is the activation of the meaning that causes
the perceptual sensations. Thus, the associative
semantic networks, which play a pivotal role for nonsynaesthetic associations, play a central functional
role also in synaesthesia: The usual concept-toconcept associations are extended to the unusual
concept-to-percept associations.

These conclusions suggest that we should adjust our

understanding of the nature of synaesthesia, which
may begin already with the definition of the
phenomenon. In light of the presented evidence, it
appears incorrect to define synaesthesia as a
phenomenon of intermixed, unified, or crosswired senses. Currently, such definitions of the
phenomenon of synaesthesia predominate in the
literature, but would obviously have to be replaced
by new, more accurate definitions. These would have
to involve the terms semantic, meaning or
concepts. Here is one proposal: Synaesthesia is a
phenomenon in which a mental activation of a
certain concept or idea is associated consistently
with a certain perception-like experience.
Importantly, the issue of the nature of synaesthesia
still cannot be considered fully settled. A number of
questions remain open. For example, it will be
important to extend the studies to forms of
synaesthesia other than grapheme-color. Also, the
present studies cannot be taken as evidence of nonexistence of low-level synaesthesia, i.e. a form that is
implied literally by the word synaesthesia, and could
be also labeled as synaesthesia literal. Besides being
virtually impossible to prove non-existence of
anything, there are other reasons to believe that
synaesthesia literal may exist in a form that is
different from the one studied typically. One
possibility is that drug-induced synaesthesia is a true
low-level phenomenon. In that case, high-level
semantic forms of synaesthesia (e.g. grapheme-color)
and temporary, drug-induced synaesthesia may be
two different, unrelated phenomena that rely on
different mechanisms and are hence, incorrectly
labeled the same. If this turns out to be true, it will be
no longer justified to refer to one in the context of the
other.
Finally, in light of the present conclusions, it
becomes obvious that the very name of the
phenomenon, synaesthesia, is inaccurate and
misleading. The question is then, is there a better
name for the phenomenon? To be consistent with the
tradition, one can ask the following: How would the
ancient Greek philosophers name this phenomenon if
they also too into consideration the semantic nature
of the inducers? One alternative is the word
ideaesthesia, which is a combination of two ancient
Greek words, one for concept, idea, and the other
for sensation, aisthesis. In translation, ideaesthesia
means sensing concepts or perceiving meaning. This
sends a considerably different message than does
union of senses. Thus, given the available empirical
evidence, the word ideaesthesia describes the
discussed phenomenon much more accurately than
the word synaesthesia.

Danko Nikoli. Received both a diploma in Psychology

(1994) and a bachelor's degree in Civil Engineering (1992)
at the University of Zagreb, Croatia. Ph.D. (1999) received
from a Department of Psychology, at the University of
Oklahoma, USA. Currently, he is a group leader at the
Max-Planck Institute for Brain Research, Frankfurt am
Main, Germany. Scientific interests: Electrophysioloy of
visual cortex, Visual working memory and Attention,
Synaesthesia, Applied scientific epistemology.

Acknowledgements
This work is partially
Foundation.

supported

by

Hertie

REFERENCES
DIXON M.J., SMILEK D., CUDAHY C., MERIKLE P.M.
(2000), Five plus two equals yellow. in Nature,
Vol. 406, pp. 365.
DIXON M.J., SMILEK D., DUFFY P.L., ZANNA M.P.,
MERIKLE P.M (2006), The Role of Meaning in
Grapheme-Colour Synaesthesia, in Cortex, Vol.
42, pp. 243-252.
EDQUIST J., RICH A.N., BRINKMAN C., MATTINGLEY
J.B. (2006). "Do synaesthetic colours act as
unique features in visual search?", in Cortex,
Vol. 42, pp. 222231.
HUBBARD E.M., ARMAN A.C., RAMACHANDRAN
V.S., BOYNTON G.M (2005), Individual
differences among grapheme-color synesthetes:
Brain-behavior correlations, in Neuron, Vol. 45,
pp. 975985.
MAYER J., BITTNER R., LINDEN D., NIKOLI D.
(2007),
Attentional
demand
influences
strategies for encoding into visual working
memory, in Advances in Cognitive Psychology,
Vol. 3, pp. 429-448.
MROCZKO A., METZINGER T., SINGER W., NIKOLI
D. (2008), "Immediate transfer of synaesthetic
concurrents to novel inducers", in The 4th
Annual General Meeting and Conference of the
UK Synaesthesia Association, Department of
Psychology, University of Edinburgh, March 2930.
NIKOLI D., LICHTI P., SINGER W. (2007a), Color
opponency in synesthetic experiences in
Psychological Science, Vol. 18, pp. 481-486.
NIKOLI D., LICHTI P., SINGER W. (2007b),
Synaesthesia does not enable extraordinary
speed of visual search, in UK Synaesthesia
Association Meeting, Oxford, March 24-25.
NUNN J.A., GREGORY, L.J., BRAMMER M., WILLIAMS
S.C.R, PARSLOW D.M, MORGAN M.J. MORRIS
R.G., BULLMORE E.T., BARON-COHEN S., GRAY
J.A. (2002), Functional magnetic resonance
imaging of synaesthesia: Activation of V4/V8 by
spoken words, in Nature Neuroscience, Vol. 5,
pp. 3715.
PALMERI T.J., BLAKE R., MAROIS R., FLANERY M.A.,
WHETSELL W. (2002) The perceptual reality of
synesthetic colors, in Proceedings of the

National Academy of Sciences, USA, Vol. 99, pp.

41274131.
RAMACHANDRAN V.S., HUBBARD E.M. (2001),
Psychophysical investigations into the neural
basis of synaesthesia, in Proceedings of Royal
Society of London Series B-Biological Sciences,
Vol. 268, pp. 979983.
RICH A.N., MATTINGLEY J.B. (2003), The effects of
stimulus competition and voluntary attention on

colour-graphemic synaesthesia, in NeuroReport,

Vol. 14, pp. 17931798.
SAGIV N., ROBERTSON L. (2006) Does binding of
synesthetic color to the evoking grapheme
require attention?, in Cortex, Vol. 42, pp. 232
242.
SIMNER J., WARD J. (2006), Synaesthesia: The taste
of words on the tip of the tongue, in Nature,
Vol. 444, pp. 438.